Paradigms of Oxygen Therapy in Critically Ill Patients
Paradigms of Oxygen Therapy in Critically Ill Patients
Paradigms of Oxygen Therapy in Critically Ill Patients
DOI: 10.21767/2471-8505.100062
Tel: 351935848475
Received: December 09, 2016; Accepted: December 20, 2016; Published: December
27, 2016
Citation: Carvalho M, Soares M, Machado
HS. Paradigms of Oxygen Therapy in
Critically Ill Patients. J Intensive & Crit Care
Introduction 2017, 3:1.
minimal atelectasis but the time to desaturation was significantly increased inter-cellular adhesion molecules, neutrophil recruitment,
shortened compared to 100% FiO2. 8-iso PGF2 alpha, nitrous oxide, and synthesis of collagen-I and
monocyte chemotactic protein 1 [14,15].
It has been proposed that high FiO2 reduces incidence of surgical
wound infection [8-10]. Surgical-wound infections (SWI) are Davis et al. [16] evaluated early reversible changes in human
common and can be a serious problem with implications in alveolar structures induced by hyperoxia. The authors evaluated
morbidity and mortality. The oxidative process is essential in the 14 normal subjects by broncho alveolar lavage before and
defense against pathogens. Greif et al. [8] included 500 patients immediately after they had breathed more than 95% oxygen for
proposed to colorectal surgery and showed that exposure to 80 16.7 ± 1.1 h. They concluded that although some of the effects
FiO2% during anesthesia and 2 h post op reduce the incidence of of oxygen exposure for 17 h were reversible, hyperoxia for this
SWI when compared to exposure of 30% FiO2. short time period can decrease the structural or functional
barriers that normally prevent alveolar-capillary leak and induce
In 2009, a meta-analysis supported these results by showing
processes that may culminate in fibrosis of the alveolar wall.
that supplemental perioperative oxygenation is beneficial
in preventing SWI in patients undergoing colorectal surgery Regarding toxicity induced by oxygen, Machado et al. [17]
[9]. More studies may be needed to determine if the use of conducted an experimental prospective randomized study
supplemental oxygen is beneficial for patients undergoing other to evaluate lung inflammation with oxygen supplementation
types of surgery. The use of 100% FiO2 during perioperative in tracheotomized spontaneously breathing rabbits. Rabbits
period seems to increase antimicrobial and pro-inflammatory were randomly assigned to 4 groups: groups 1 and 2 were
responses of the alveolar macrophages during anesthesia. submitted to
0.5 L/min oxygen supplementation, for 20 or 75
min, respectively; groups 3 and 4 were left at room air for 20
Kotai et al. [11] suggested that these alterations may be beneficial
or 75 min. The results showed that oxygen supplementation
for immune defense in lung tissue.
in spontaneous breathing is associated with an increased
The effect of high concentration of FiO2 on postoperative nausea inflammatory response when compared to breathing normal
and vomiting (PONV) is still not clarified as previous studies had room air, and this inflammatory response was mainly constituted
divergent results. In 2013, a meta-analysis including 11 studies by polymorphonuclear cells [17].
regarding the impact of high FiO2 and PONV showed a weak
Another physiologic change that can be induced by high oxygen
beneficial effect on nausea [10].
inspired fraction is the development of absorption atelectasis.
Despite the increase of atelectasis formation with exposure to Atelectasis may develop with pre-oxygenation with 100% FiO2
high FiO2, there are studies showing that this have no impact [18]. This is a major issue and needs a special intervention,
in pulmonary function. Besides that, it has been suggested that namely the use of positive end expiratory pressure (PEEP) to
there is no difference in the incidence of pulmonary complications maintain the lungs open [18] as even recent investigation has
between those who were exposed to high FiO2 and low FiO2 shown that if this is not instituted to the patient, the lung will
during perioperative period [12]. certainly be poorly aerated [19].
Finally, the work of Lambertsen and his colleagues at the Hedenstierna [20] has shown that patients whose anesthesia
University of Pennsylvania has shown the relationship between was induced during pre-oxygenation with 100% O2 for 3 min had
inspired oxygen partial pressures and the time for the appearance atelectasis, on average, in 10% of the lung area (corresponding
of pulmonary and central nervous system toxicity symptoms. It to 15-20% of lung tissue). Pre-oxygenation with 80% O2 for a
seems that neurologic toxicity occurs only when oxygen exposure similar time caused much less atelectasis, with a mean of 2%.
happens at high pressure and pulmonary toxicity appears when Reducing inspired oxygen to 60% during pre-oxygenation almost
exposure to high FiO2 oxygen is prolonged in time [13]. eliminated atelectasis and nothing was observed with 30% O2.
Also, the duration of pre-oxygenation with 100% oxygen can
Cons affect the amount of atelectasis that is produced. Thus, pre-
oxygenation with 100% O2 for 6 min resulted in a mean of 6%
It is very important to carefully decide on the level of inspired atelectasis, whereas 100% O2 for shorter time of ventilation (3
fraction that should be used during Oxygen is needed in order to min) resulted in a little more than 2%. Finally, 30% oxygen for 3
assure patient safety in critical moments, but excessive amounts min resulted in barely visible atelectasis of around 0.3-0.4%. The
of oxygen inspired fractions might induce physiologic lesions. focus anesthetic regime should be to deliver a patient with no
It is well recognized and confirmed by recent publications that atelectasis to the postoperative ward and to keep the lung open.
hyperoxia may induce pulmonary toxicity and cell damage
mediated by radical oxygen species (ROS). High doses of oxygen In Stroke Patients
have also been associated to target-organ damage, mainly brain Few studies have addressed supplemental oxygen therapy in
and heart, decreased cerebral and coronary blood flow, systemic stroke patients and the results are controversial. Rønning et al.
vasoconstriction and decreased cardiac output. [21] performed a study involving 550 stroke patients who were
High oxygen inspired fractions have been correlated with randomly assigned between two groups: to a group that received
pulmonary toxicity induced by several mechanisms, including supplemental oxygen treatment (100% atmospheres, 3 L/min)
for 24 h (n=292) or to a control group, which did not receive for the increase in mortality rates over ascending PaO2 ranges
additional oxygen. Results showed that there was no statistically (p<0.0001). The results of this multicenter study showed a linear
significant difference in one year survival between groups, but dose-dependent relationship between supranormal oxygen
patients with less severe strokes (Scandinavian Stroke Scale >40) tension and relative risk of in-hospital mortality, but no evidence
had a higher and significant survival rate in the control group (OR supporting a single threshold for harm from supranormal oxygen
0.45; 95% CI 0.23 to 0.90; p=0.023). tension [28].
More recently, Roffe et al. [22] conducted a randomized clinical
trial in 289 patients who suffered a stroke and were given
Acute Myocardial Infarction
supplemental oxygen therapy for 72 h after admission on the Oxygen is widely used in people with acute myocardial infarction
local stroke unit. They found a slight short-term improvement (AMI) although evidence of its benefit is controversial. Many
in the oxygen group regarding NIHSS at end of one week, but systematic reviews have concluded that there is insufficient
their findings may have been biased by the imbalance of stroke evidence to know whether oxygen reduces increases or does not
severity between the two groups. have an effect on heart ischemia or infarct size.
Finally, The Stroke Oxygen Pilot Study did not find long-term The rationale for supplemental oxygen in AMI is that it may
significant differences in outcome among stroke patients improve the oxygenation of the ischemic myocardial tissue and
submitted to supplemental oxygen therapy compared to reduce ischemic infarct size, morbidity and mortality. However,
standard of care [23]. potentially harmful mechanisms include the paradoxical effect
of oxygen in reducing coronary artery blood flow and increasing
There is a general consensus in what concerns the need for larger
coronary vascular resistance [29,30] reducing stroke volume
studies in this field, but the best current evidence states that
and cardiac output and the consequent reperfusion injury from
oxygen therapy should not routinely be used in stroke patients
increased oxygen free radicals [31,32].
and it might be deleterious in patients with minor strokes.
In 2013, a Cochrane meta-analysis reviewed 4 articles where
In Resuscitation Following Cardiac oxygen supplemental therapy was used in over 400 non-hypoxic
Arrest patients. When the data were pooled, twice as many people on
oxygen died when compared to the group that was given air.
A primary contributor to death and disability in resuscitated Meta-analysis for mortality in participants with confirmed AMI:
patients is the anoxic brain injury that typically follows a severe risk ratio (RR) 2.11 (95% confidence interval (CI) 0.78 to 5.68);
ischemia/reperfusion insult. To date, therapeutic hypothermia is meta-analysis for mortality in an intention-to-treat population
the only proven treatment for anoxic brain injury after cardiac (including those who did not have AMI): RR 2.05 (0.75 to 5.58)
arrest [24-26]. However, recent evidence suggests that, in [33].
addition to body temperature, the amount of oxygen delivered
However, due to the low number of deaths in all four studies (17
to the brain after reperfusion may represent an important and
in total), the authors recognized that evidence is lacking power
modifiable factor in post resuscitation care. It is common to
and results may have been influenced by chance. As of such,
deliver supplemental oxygen in maximum concentrations during
current evidence does not support or clearly refutes the use of
the resuscitative period, but this strategy may not be innocuous.
supplemental oxygen therapy in non-hypoxic AMI patients.
Numerous laboratory investigations have identified a paradox
relative to oxygen delivery to the injured brain. One large-scale
clinical study has examined the relationship between excessive
Neonatal Resuscitation
supplemental oxygen and outcome in post resuscitation patients Up to 10% of newborns require some degree of resuscitation,
[27]. Post resuscitation hyperoxemia exposure, defined as a ranging from simple stimulation to assisted ventilation [34]. The
partial pressure of arterial oxygen (PaO2) 300 mm Hg, was found aim of resuscitation is to prevent not only neonatal death but also
to be an independent predictor of in-hospital death. The mortality the adverse long-term neurodevelopmental squeals associated
in patients with hyperoxemia was even higher than the mortality with birth asphyxia [35].
for patients with hypoxemia. Furthermore, hyperoxemia was Recommendations point out 100% oxygen as the gas to use during
associated with lower likelihood of independent functional status neonatal resuscitation. Nonetheless, many experts challenge this
among patients who survived to hospital discharge [27]. practice on the basis that little evidence exists to support it [35].
Kilgannon et al. [28] designed a multicenter cohort study using Increasing work of breathing and the increased generation of
the Project IMPACT database (intensive care units at 120 US oxygen free radicals are amongst the potential adverse effects of
hospitals) to better define the relationship between supranormal supplemental oxygen [36].
oxygen tension and outcome in post resuscitation patients. The Cochrane meta-analyses of five studies comparing 100% oxygen
proportion of in- hospital deaths was 41% for patients with a versus room air in neonatal resuscitation showed a reduction
PaO2 in the range of 60 to 99 mm Hg and rose to 65% for patients in mortality in the room air group, with no increase in future
with a PaO2 in the range 400 mm Hg. The authors observed no complications (including long-term neurodevelopmental outcome)
clear single threshold of increased risk of death over ascending [35]. However, long-term follow-up (18-24 months) was only
PaO2 ranges. In addition, they found a significant linear trend obtained in one study and the proportion of eligible patients
was under 70%. Also, up to 25% of patients assigned to the room usual oxygenation strategy, while being effective in reducing
air group received back-up 100% oxygen therapy. The authors exposure to hyperoxia [38,39].
concluded that there is no evidence to recommend using room
There is still insufficient evidence regarding the safety of arterial
air over 100% oxygen or vice versa and that future research is hyperoxia in critically ill patients. Most of the existing studies
required in this field. are observational investigations with highly heterogeneous
characteristics and inconsistent results. Randomized controlled
In Critical Care trials are lacking.
Oxygen administration is the most widely prescribed therapy Arterial hyperoxia may be associated with higher mortality in
in critically ill patients and frequently represents a life-saving some critically ill patient subsets (post-cardiac arrest, stroke and
intervention. Clinical data regarding the relationship between traumatic brain injury).
arterial hyperoxia and outcome are contradictory. The question
whether exposure to supranormal arterial O2 tensions (PaO2) is While a liberal use of oxygen may provide a margin of safety
safe in critically ill patients remains unanswered. against hypoxia, a more conservative approach might reduce
potentially harmful exposure to excessive FiO2, hyperoxemia and
In 2014, Damiani et al. performed a systematic review and meta- tissue hyperoxia.
analysis of studies describing the relationship between arterial
hyperoxia and mortality in critically ill patients [37]. The majority Conclusion
of studies that have explored the relationship between arterial The selection of optimum arterial oxygenation goals is essential
hyperoxia and mortality in critically ill patients are retrospective if cellular hypoxia and unnecessarily excessive oxygenation (and
observational investigations, with only one prospective before- ventilation) are to be avoided. It is imperative to balance the risks
after study, which supported the use of a more conservative associated with hypoxemia and hyperoxemia forms during daily
strategy. The authors concluded that hyperoxia exposure may assessment of patients.
be associated with mortality in patient subsets (post-cardiac
The traditional clinical approach to oxygen therapy has been
arrest, stroke and traumatic brain injury). However, they referred
to prioritize the avoidance of hypoxemia while being relatively
that these results must be interpreted carefully given the
tolerant of hyperoxemia. The possibility that “too much” oxygen
heterogeneity in criteria used for defining hyperoxia exposure may be as harmful as “not enough” should lead to a pragmatic
and a significant inconsistency between study findings. rethinking of the practice of oxygen administration.
Panwar et al. conducted a multicenter randomized controlled Perhaps, the most prudent and advisable practice when oxygen
trial to determine whether a conservative oxygenation strategy needs to be used, will be to think of it as a lifesaving drug with
is a feasible alternative to a liberal oxygenation strategy among its undeniable advantages and uses, and at the same time, a gas
ICU patients requiring invasive mechanical ventilation (IMV). that has to be handle with special care, thought that hypoxia may
At four multidisciplinary ICUs, 103 adult patients deemed likely be managed by several means (PEEP, recruitment maneuvers)
to require IMV for ≥ 24 h were randomly allocated to either a before the decision of a high- inspired fractions of oxygen should
conservative oxygenation strategy with target SpO2 of 88-92% be always in our minds.
or a liberal oxygenation strategy with target SpO2 of ≥ 96%. The Given the widespread use of O2 therapy in critical care, clinicians
study findings support the feasibility of delivering conservative should be aware of the potentially deleterious effects of excessive
oxygen therapy in patients on invasive MV as alternative to the O2 administration.
14 Huang D, Fang F, Xu F (2016) Hyperoxia induces inflammation and 28 Kilgannon JH, Jones A, Parillo J, Dellinger P, Milcarek B, et al. (2011)
regulates cytokine production in alveolar epithelium through Relationship between supranormal oxygen tension and outcome
TLR2/4-NF-kB-dependent mechanism. Eur Rev Med Pharm Sci 20: after resuscitation from cardiac arrest. Circulation 123: 2717-2722.
1399-1410. 29 McNulty PH, King N, Scott S, Hartman G, McCann
J, et al. (2005)
15 Kotani N, Hashimoto H, Sessler D, Muraoka M, Hashiba E, et al. Effects of supplemental oxygen administration on coronary blood
(2000) Supplemental intraoperative oxygen augments antimicrobial flow in patients undergoing cardiac catheterization. Am J Phys, Heart
and proinflammatory response of alveolar macrophages. Anesth 93: & Circ Phys 288: H1057-H1062.
3-5. 30 McNulty PH, Robertson BJ, Tulli MA, Hess J, Harach LA, et al. (2007)
16 Bruce DW, Rennard S, Bitterman P, Crystal R (1983) Early reversible Effect of hyperoxia and vitamin C on coronary blood flow in patients
changes in human alveolar structures induced by hyperoxia. N Eng J with ischaemic heart disease. J Appl Physiol 102: 2040-2045.
Med 309: 878-883. 31 Milone SD, Newton GE, Parker JD (1999) Hemodynamic and
17 Machado H, Nunes C, Sa P, Couceiro A, Silva A, et al. (2014) Increased biochemical effects of 100% oxygen breathing in humans. Canadian
lung inflammation with oxygen supplementation in tracheotomized Journal of Physiol and Pharmacology 77: 124-130.
spontaneously breathing rabbits: An experimental prospective 32 Rousseau A, Bak Z, Janerot-Sjoberg B, Sjoberg F (2005) Acute
randomized study. BMC Anesth 14: 86. hyperoxemia-induced effects on regional blood flow, oxygen
18 Hedenstierna G (2014) Effects of anaesthesia on ventilation/ consumption and central circulation in man. Acta Physiologica
perfusion
matching. Eur J Anaesthesiol 31: 447-449.
Scandinavica 183: 231-240.
33 Cabello JB, Burls A, Emparanza JI, Bayliss S, Quinn T (2013) Oxygen 37 Damiani E, Adrario E, Giradis M, Romano R, Pelaia P, et al. (2014)
therapy for acute myocardial infarction.
Cochrane Database Syst Arterial hyperoxia and mortality in critically ill patients: A systematic
Rev 21: CD007160. review and meta-analysis. Crit Care 18: 711.
34 Saugstad OD (1998) Practical aspects of resuscitating new-born 38 Panwar R, Hardie M, Bellomo R, Barrot L, Eastwood G, et al. (2016)
infants. Eur J Ped 157: S11-S15. Conservative versus liberal oxygenation targets for mechanically
35 Tan A, Schulze AA, O’Donnell CPF, Davis PG (2005) Air versus oxygen ventilated patients - A pilot multicenter randomized controlled trial.
for resuscitation of infants at birth. Cochrane Database Syst Rev 2: Am J Respir Crit Care Med 193: 43-51.
CD002273.
39 Machado HS (2016) O2 and N2O: Between the rock and the hard
36 Mortola JP, Frapell PB, Dotta A (1992) Ventilatory and metabolic place. J Anesth Clin Res 7: 5.
response to acute hyperoxia in newborn. Am Rev Resp Dis 146: 11-15.